Nuclear DNA influences variation in mitochondrial DNA

Whole genomes from hundreds of thousands of people reveal new complexity in how the nuclear and mitochondrial genomes interact, which may influence how cells produce energy.

Graphic showing a nucleus and mitochondrion communicating within a cell.
Credit: Sonja Vasiljeva, Broad Communications

The energy-producing machines inside cells, called mitochondria, have their own DNA that’s passed down from mother to child. The mitochondrial genome, which encodes just 13 proteins, is smaller and comparatively less well studied than the genome in the cell’s nucleus, even though mutations in mitochondrial DNA can cause a number of rare diseases. 

Now a new study of both mitochondrial and nuclear genomes from hundreds of thousands of people may change how scientists think about the mitochondrial genome and how it interacts with the nuclear genome. The findings could inform future studies of how this crosstalk helps mitochondria power the cell, and shed light on when they cause disease.

Scientists have long known that cells can have hundreds, even thousands, of copies of the mitochondrial genome, and that this “copy number” can vary widely from one cell type to the next. There is also much variation in the sequences of all that mitochondrial DNA in a single cell, called heteroplasmy, which researchers had previously linked to rare inherited mitochondrial disease. 

The new study from scientists at the Broad Institute of MIT and Harvard has shown that the number of copies of the mitochondrial genome is a trait that varies from person to person and is controlled by the nuclear genome. The researchers also found that heteroplasmy is also influenced by mutations in the nuclear genome and is pervasive even among healthy people. 

The findings, published today in Nature, suggest that scientists can now quantitatively analyze copy number and heteroplasmy in mitochondria DNA in future studies to learn how the nuclear genome regulates levels of mitochondrial DNA across cells and tissues, how this relationship has evolved, and the connection between mitochondrial DNA and disease. The work is the first of its kind of this size to study both nuclear and mitochondrial DNA from humans of diverse ancestries to date.

The research team includes co-senior author Vamsi Mootha, an institute member at the Broad, investigator at the Howard Hughes Medical Institute, investigator in the department of molecular biology at Massachusetts General Hospital; co-senior author Benjamin Neale, an institute member and co-director of the Program in Medical and Population Genetics at the Broad; and first author Rahul Gupta, an MD-PhD student jointly advised by Mootha and Neale.

“We’ve long been interested in the question of how variation in the nuclear DNA influences mitochondrial DNA — it is so exciting to be able to finally answer this long standing problem,” Mootha said. “The whole project was a perfect storm of whole genome sequence availability, statistical power, and the right people with the right skills.”

“Mitochondrial DNA is a very ancient molecule and has co-evolved with the nuclear genome,” Gupta said. “Our work provides a window into how these two genomes coexist and might influence each other to facilitate one of the most important processes in our cells: the pathway that generates energy.”

Mitochondrial genomics

The mitochondria rely on more than 1,000 proteins to function, most of which are encoded by DNA in the nucleus, so scientists have long suspected that nuclear DNA can influence mitochondrial DNA but they didn’t know how. For more than a decade, Mootha wanted to study these questions, but the team didn’t have enough human whole genome sequences to be able to do this thoroughly. 

Then, in 2022, the UK Biobank and the National Institutes of Health’s All of Us program published whole genome sequences, which contained often-discarded information on mitochondrial DNA sequences. Gupta and colleagues jumped on the opportunity to finally tackle these questions and gathered sequences from more than 250,000 individuals across six ancestry groups. 

“These large-scale biobanks with an open science philosophy are transforming what is possible in biomedical research and are accelerating the pace of discovery across the community,” Neale said.

Building on a workflow previously developed by Broad Institute’s Sarah Calvo, a computational biologist in Mootha’s lab, Gupta constructed a computational pipeline to efficiently analyze variation in mitochondrial DNA within the whole genome sequences. They studied the copy number of mitochondrial DNA, mutations present within the mitochondrial DNA itself, and the places in the nuclear genome influencing these traits. 

They found that copy number declines consistently with age, which is in line with previous studies. But the researchers were surprised to discover, once they accounted for confounding factors, that copy number is not associated with most common diseases. This is contrary to previous studies, which did not account for these factors. The researchers found that copy number is influenced by the nucleus and pinpointed 92 locations in the nuclear genome that regulate copy number, many of which have been linked to rare genetic disorders related to mitochondrial DNA maintenance.

Mitochondrial messages

The team also found that heteroplasmy follows two patterns over a person’s life. Single-letter changes called single nucleotide variants accumulated with age, particularly after 70. In contrast, the number of short insertions or deletions — called “indels” — in mitochondrial DNA, which are common, stay the same throughout life and are passed down from mother to child. The researchers identified 42 different locations in the nuclear genome, including some involved in replication and maintenance of mitochondrial DNA, that influence the levels of these mitochondrial indels across the human population. 

“Almost every person has these mitochondrial indels, and the relative amount of those variants is influenced by variation in the nuclear genome, and that’s profound,” Mootha said. “We believe that this phenomenon represents ‘genetic interactions’ between the nuclear genome and the mitochondrial DNA that are crucial for their joint success. These influences had long been theorized to exist but we’ve now been able to pinpoint many convincing instances.”

The work also suggests that the nuclear genome favors the replication of certain mitochondrial DNA variants. Mootha said that insight could one day help improve mitochondrial replacement therapies, which aim to replace disease-causing mitochondrial DNA with DNA from a healthy donor. “Our work is beginning to reveal the mechanistic logic of nuclear and mitochondrial DNA matching and could have long-term implications for new therapies that aim to prevent the transmission of mitochondrial disease,” he said. 

Gupta said that further study of even more genomes could reveal more subtle effects of the nuclear DNA on the mitochondrial genome. And because the scientists found that the mitochondrial DNA mutations under nuclear control are passed down from mother to child, Gupta added that he’d like to use family structure to better understand how heteroplasmy changes over generations.

“A lot of people think about human genetics being applied to GWAS [genome-wide association studies] for important traits like height or type 2 diabetes,” Gupta said. “But it’s really cool to be able to apply it to gain insights into very basic cellular phenomena as well — and now we have some really exciting candidates we can use to dissect mitochondrial biology.”


This work was supported by the Howard Hughes Medical Institute.

Paper cited

Gupta R et al. Nuclear genetic control of mtDNA copy number and heteroplasmy in humans. Nature. Online August 16, 2023. DOI: 10.1038/s41586-019-0000-0.